Foodinformatics

244 K. Martinez-Mayorga et al. As detailed in Chap. 1 of this book, similarity searching is at the core of chemoin- formatics, and multiple articles are published frequently on this topic. As expected, commercial software as well as programs developed by various research groups are available. For instance, ChemAxon, mentioned above, is a chemoinformatics plat- form and has a robust implementation of similarity searching methods. MOE and Schrödinger have also implemented structural similarity methods. Online resources are also available. For instance, MOLPRINT 2D and SEA (listed in Table 9.2) pro- vide for similarity searching, the former for ligands and the latter for proteins. Finally, an important commonly pursued goal is the prediction of bioactivity as well as ADME/tox properties. Physicochemical properties and bioactivities, and ADME/tox properties can be calculated through online services such as VCCLab, PASS, FAF-Drugs. Table 9.2 provides the corresponding websites, for more information. Being a major concern, a number of initiatives are dedicated to food safety is- sues. Some programs are maintained by, or in cooperation with, universities while others are consortia involving, in many cases, governments. Some of them are: the Centers for Disease Control [36], the Food Safety Research Consortium [37], ComBase [38], and Bits [39]. Developed and/or maintained by universities are: FareMicrobial [40], and the Center for Food Safety [41]. In addition, other online services are specifically focused on food information (food informatics). In these cases, the information contained is not necessarily di- rectly related to chemical structures, though it does illustrate the versatility of and need for using information technology to excel on tasks having a direct impact on health and well-being through food and nutrition. Vision Software assists the organization, storage and use of information, data and knowledge for food and nutrition-related problem solving and decision mak- ing. One direct application is in the area of diets for hospitals, hotels, etc. [42]. A related novel piece of work (albeit somewhat controversial) concerns so-called food pairing theory. The hypothesis is that a pair of ingredients which share many flavor compounds accompany each other better than those that do not, e.g., bacon and cheese, asparagus and butter, and chocolate and blue cheese. This food pairing concept is useful to understand and further develop culinary practice [43]. The Food & Biobased Research Company (Wageningen UR) has five major projects. One of them, called Food Informatics, focuses directly on food research. This project is conducted in cooperation with the Top Institute-Food & Nutrition (TIFN), Unilever Research, TNO Quality of Life research center, and Friesland Foods. Their focus is on the modeling of knowledge-intensive processes and the development of corresponding applications. Based on ontologies, Top et al. have focused on methods and tools for extracting knowledge in the food industry domain [44]. For example, using this approach, they have developed an on-line system for searching the properties and practical applications of five natural antimicrobial preservatives and their relationships to a large number of microbes and food types [44]. Another example consists of how these methods can help as decision support in the fruit and vegetable supply chain [45, 46].

9  Software and Online Resources: Perspectives and Potential Applications 245 Another online service is called Nutrition Informatics. Nutrition informatics is defined as “the effective retrieval, organization, storage, and optimum use of information, data, and knowledge for food and nutrition-related problem solving and decision making. Informatics is supported by the use of information standards, information processes, and information technology.” As part of the Academy of Nutrition and Dietetics, the Nutrition Informatics group provides a service to regis- tered members. It is the intersection of information, nutrition, and technology. The hugely data-rich food-related information includes food/nutrient analysis tables. This provides registered dietitians web-based tools, allowing them to use their knowledge and skills more efficiently in making dietary recommendations [47]. Another service that can be classified in this category is that provided by the O’Neill Institute for National & Global Health Law. It is a free online database of law, from around the world, relating to health and human rights. The database offers an interactive, searchable, and fully indexed website of case law, national constitu- tions, and international instruments [48]. 9.3 Perspectives and Potential Applications While the exploitation of chemical information in the food chemistry field is still emerging, this has already proven to constitute a useful approach, as illustrated through several examples described in the second section of this book and as also reported in the literature elsewhere. The use of similarity to compare and explore food-related databases, described in Chap. 1 and exemplified in Chap. 3 clearly demonstrate the applicability of these methods and alludes to exploring other applications; for example, expanding the studies to diseases, methodologies, and databases beyond those explored in Chap. 3. In addition, methods such as artificial neural networks proved useful when exploring the effects of foods on cancer cell growth, suppression activity, antivi- ral activity and antioxidant stress activity; this, of course, suggests exploring other diseases. The use of information technology in the food and beverage field is not limited to chemical structures. In fact, data mining is widely used to collect, organize, analyze, and archive diets in hospitals and restaurants, just as it is applied to chemical struc- tures in the area of chemical information. The theory behind the methods devised to perform these tasks is general and can be applied to datasets regardless of origin. Therefore, the software employed in drug discovery can readily be used or adapted to food chemical applications. In the same way that new concepts and methodologies are developed on an almost daily basis and reported in chemical information jour- nals, the food chemical information field can be expected to grow significantly during the coming decade. Taking into account the knowledge and applicability of chemical information devoted to drug discovery, and considering the inherent complexity of the food chemistry field, it is also expected that concepts now devel- oped in the food chemical information area will feed back into the drug discovery

246 K. Martinez-Mayorga et al. arena. This can be exemplified by the well-known complexity of odor perception, where multiple odor receptors are activated by multiple ligands, ultimately to pro- duce specific percepts. The multiple receptor/multiple ligand notion is central to the polypharmacology concept that is gaining attention nowadays. The idea of multiple target responses is neither new nor unexpected, however, only relatively recently is the paradigm change from single target to multitarget being recognized [49]. This interplay between drug discovery and food chemical information is not only promising but has also proven to be useful and may yet further expand our knowl- edge and boost our creativity in developing new methods to deal with complex multivariable systems. In the light of the discussion above, there is clearly a need for professionals with skills in information technology and a strong background in food chemistry. To fill this need, it will be necessary to explore the suitability of various chemoinformatic methods, selecting and developing the most useful candidates, and to design appro- priate programs and courses in universities. This has been the path of the chemical information field, but with emphasis in drug discovery. For example, chemoin- formatics courses leading to a Masters in Science have been implemented at the University of Sheffield, the University of Manchester and Indiana University. How fast we respond to this need will have an impact not only on the development of the field but also on how we take advantage of the emerging field of food informatics. Acknowledgments  K.M-M. thanks the Institute of Chemistry-UNAM and DGAPA-UNAM for funding (PAPIIT IA200513-2). The authors also wish to thank Robertet Flavors for permission to publish this chapter. References ­  1. Martinez-Mayorga K, Medina-Franco JL (2009) Chemoinformatics—applications in food chemistry, vol. 58. Elsevier, Burlington   2. Martínez-Mayorga K, Peppard TL, Yongye AB, Santos R, Giulianotti M, Medina-Franco JL (2011) Characterization of a comprehensive flavor database. J Chemometr 25:550–560  3. Sprous DG, Salemme FR (2007) A comparison of the chemical properties of drugs and FEMA/FDA notified GRAS chemical compounds used in the food industry. Food Chem Toxicol 45:1419–1427   4. Medina-Franco JL, Martínez-Mayorga K, Peppard TL, Del Rio A (2012) Chemoinformatic analysis of GRAS (Generally Recognized as Safe) flavor chemicals and natural products. PLoS ONE 7:e50798   5. Femaflavor. http://www.femaflavor.org  6. Hallagan JB, Hall RL (1995) FEMA GRAS—a GRAS assessment program for flavor ingredients. Regul Toxicol Pharm 21:422   7. Hallagan JB, Hall RL (2009) Under the conditions of intended use—new developments in the FEMA GRAS program and the safety assesment of flavor ingredients. Food Chem Toxicol 47:267–278  8. http://www.rifm.org/index.php  9. http://www.iofi.org 10. http://eur-lex.europa.eu/JOHtml.do?uri=OJ:L:2012:267:SOM:EN:HTML 11. https://webgate.ec.europa.eu/sanco_foods/main/?event=display

9  Software and Online Resources: Perspectives and Potential Applications 247 12. http://eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri=OJ:L:2012:267:FULL:EN:PDF 13. http://www.accessdata.fda.gov/scripts/fcn/fcnNavigation.cfm?filter=&sortColumn=&rpt=ea fusListing&displayAll=false#1 14. http://www.usp.org/food-ingredients/food-chemicals-codex 15. http://www.leffingwell.com/flavbase.htm 16. Leffingwell J, Leffingwell D (2014) Perfumer Flavorist 39:26–37 17. http://www.vcf-online.nl/VcfHome.cfm 18. http://www.leffingwell.com/baciseso.htm 19. http://dir.perfumerflavorist.com/main/login.html;jsessionid=9EC896163AA3A88037DD0B C0E2CE6F65 20. http://www.flavornet.org 21. http://acree.foodscience.cornell.edu/flavornet.html 22. Arnam H, Acreeb TE (1998) Flavornet: a database of aroma compounds based on odor potency in natural products. Dev Food Sci 40:27 23. http://bioinf-applied.charite.de/superscent 24. López-Vallejo F, Peppard TL, Medina-Franco JL, Martínez-Mayorga K (2011) Compu- tational methods for the discovery of mood disorder therapies. Expert Opin Drug Discov 6:1227–1245 25. Dunkel M, Schmidt U, Struck S, Berger L, Gruening B, Hossbach J, Jaeger IS, Effmert U, Piechulla B, Eriksson R, Knudsen J, Preissner R (2008) SuperScent–a database of flavors and scents. Nucleic Acids Res 37:D291–D294 26. http://www.thegoodscentscompany.com/index.html 27. http://www.phenol-explorer.eu/ 28. Wiener A, Shudler M, Levit A, Niv MY (2011) BitterDB: a database of bitter compounds. Nucleic Acids Res 40:D413–D419 29. http://bitterdb.agri.huji.ac.il/bitterdb 30. http://senselab.med.yale.edu 31. http://www.sigmaaldrich.com/chemistry/flavors-and-fragrances.html 32. http://frutaromfandfingredients.com/ingredients4u/Templates/showpage.asp?DBID=1&LN GID=1&TMID=84&FID=517 33. Irwin JJ, Shoichet BK, Mysinger MM, Huang N, Colizzi F, Wassam P, Cao Y (2009) Automated docking screens: a feasibility study. J Med Chem 52:5712–5720 34. Consortium TU (2014) Activities at the universal protein resource (UniProt). Nucleic Acids Res 32:W321–W326 35. Rost B, Yachdav G, Liu J (2004) The PredictProtein server. Nucleic Acids Res 32:W321–W326 36. http://www.cdc.gov/ 37. http://www.rff.org/news/features/pages/food-safety-research-consortium.aspx 38. http://www.combase.cc/index.php/en/ 39. http://bites.ksu.edu/ 40. http://foodrisk.org/exclusives/faremicrobial/ 41. http://www.ugacfs.org/ 42. http://www.vstech.com/healthcare-initiatives/food-nutrition-informatics.php 43. Ahn Y-Y, Ahnert SE, Bagrow JP, Barabási A-L (2011) Flavor network and the principles of food pairing. Sci Rep 1:196 44. Koenderink NJJP, Hulzebos JL, Roller S, Egan B, Top JL Antimicrobials on-line: concept and application for multidisciplinary knowledge exchange in the food domain. http://www. koenderink.info/nicole/pdf/Koenderink2003AFOT.pdf 45. Top JL, Rijgersberg H (2003) Modelling for decision support in the vegetable and fruit supply chain. Acta Hort 604:189–197 46. http://www.afsg.nl/InformationManagement/index.php?Itemid=0&id=21&option=com_ content&task=view 47. http://www.eatright.org/HealthProfessionals/content.aspx?id=6442471521 48. http://www.law.georgetown.edu/oneillinstitute/

248 K. Martinez-Mayorga et al. 49. Medina-Franco JL, Giulianotti MA, Welmaker GS, Houghten RA (2013) Shifting the single- target to the multi-target paradigm in drug discovery. Drug Discov Today 18:495–501 50. http://www.vcclab.org/ 51. http://www.molecularmovies.com/toolkit/ 52. http://www.chemaxon.com/ 53. http://www.chemcomp.com/ 54. http://schrodinger.com/ 55. http://www.statsoft.com/ 56. http://spotfire.tibco.com/ 57. www.miner3d.com/ 58. www.chemicalize.org/ 59. www.chemspider.com/ 60. www.elsevier.com/online-tools/reaxys 61. http://cactus.nci.nih.gov/index.html 62. http://www.uwm.edu.pl/biochemia/index.php/en/biopep 63. http://www.molprint.com/ 64. http://sea.bkslab.org/ 65. http://www.vcclab.org/lab/ 66. http://www.pharmaexpert.ru/passonline/index.php 67. http://bioserv.rpbs.univ-paris-diderot.fr/Help/FAFDrugs.html

Index A Chemical Universe Database,  84, 86, 92 Activity cliff,  28, 32, 33, 47, 55 ChemMapper, 115 ADMET,  113, 165 Chemometrics, 229 Analysis of Variance (ANOVA),  Classification methods,  214, 215 clogP, 87 MANOVA,  217, 218, 220, 222 Compound databases,  143 Aroma descriptors,  216, 218 comparison of,  57, 58 B Confidence interval circles,  215, 220 Bag-in-Box (BIB),  216, 219, 220 CREDO, 115 Bartlett’s test,  215, 218, 220, 224 Bioactive ingredients,  112, 192 D BitterDB,  83, 84, 240 Data analysis techniques,  213, 215, 228 Descriptor,  4, 7, 9, 10, 30, 36, 43, 69, 88, 220, C Canonical Variate Analysis (CVA),  214, 217 226, 237, 241 2-D LBVS,  64 of sensory data set,  220 3-D LBVS,  64 of the elemental profile data,  224, 225 3D-BCUT, 45 Canonical variates (CVs),  214, 220, 224 BCUT,  18–20, 34, 35, 44, 68 ChEMBL,  4, 43, 84, 85, 87, 113, 126 CS, 55 Chemical information,  18, 30, 34, 40, 233, Diabetes,  151, 200, 202 type 2 diabetes mellitus,  177, 178 235, 243, 245, 246 Diversity,  5, 15, 16, 19, 30, 44, 45, 58–63, 67, Chemical space (CS),  2, 5, 30 83, 86, 88, 92, 94 cell-based, 41–45 Docking,  example of,  44, 45 representations of,  44 ligand-protein, 64 protein-ligand, 167 coordinate-based,  34–37, 39, 40 DPP8 inhibitors,  182, 183, 191 derived from structural DPP9 inhibitors,  183, 191 fingerprints,  35–37, 39 DPP-IV inhibitors,  179, 182, 190, 191, non-Euclidean,  39, 40 197, 202 fragnance analogs in,  91 commercially available,  185, 187 networks, 46–56 side effects of,  187 example of,  46–49 natural products as,  192 statistical aspects of,  50–54 topologies of,  54, 55 F of flavors,  88–90 Fingerprints (FPs),  4 maps of,  89, 90 visualization of,  88, 89 atom pair,  9 of natural and food compounds,  143, 144 binary structural,  6 © Springer International Publishing Switzerland 2014 249 K. Martinez-Mayorga, J. L. Medina-Franco (eds.), Foodinformatics, DOI 10.1007/978-3-319-10226-9

250 Index extended connectivity,  9, 10 that modulate the action of PPARγ,  157, molecule-independent/directory-based,  158, 163–165 7, 8 Nearest neighbours,  91 weighted structural,  10 Flavor,  O chemical spaces of,  88–90 Octanol/water partition coefficient,  87 Odor space,  238 maps of,  89, 90 visualization of,  88, 89 P molecules, 84 Pairwise similarity measure,  91 Flavor and Extract Manufacturers Association Partial least squares regression (PLSR),  214, (FEMA),  234, 236 215, 218, 225, 228 Flavornet,  83, 84, 238 Pharmacognosy,  112, 132 Food,  Pharmacophoric features,  33, 89 Physico-chemical data,  114 additive,  112, 118, 190, 202, 236 Physicochemical properties,  167, 187, 239, databases, 233 Ftrees, 115 244 Functional food,  111, 152, 158, 164, 192, 200, Polarity,  85, 87, 89, 237, 238 Polypharmacology,  134, 138, 200, 246 202, 236 Principal component analysis (PCA),  36, 88, G 214, 217 GDB-13,  84–87, 92 Protein Data Bank,  113, 115, 242 Generally Recognized as Safe (GRAS),  115, Protein-based approach,  113 PubChem,  4, 43, 84, 114 116, 119, 135, 234 Gliptins,  185, 202 L Q Latent vectors (LVs),  215, 225 Quantitative Structure Activity Relationship Libraries,  38, 55, 57, 134, 139, 165 Ligand-based approach,  113, 119 (QSAR),  18, 19, 167 Linear combinations,  214 Quantitative Structure Property Relationship Loading plot,  214 (QSPR),  18, 19 Query molecule,  15, 16, 91, 167 M R Modeling,  Reference compound,  26, 27, 29, 42, 65, 66, homology, 113 68, 70 molecular modeling,  242 Representation,  pharmacophore,  143, 166, 167 Molecular Quantum Numbers (MQN),  88, 91 of cell-based CSs,  44 Mouthfeel descriptors,  216 self-based, 6 Multivariate analysis of variance (MANOVA) binary structural fingerprints,  6 See under Analysis of Variance vector-based, 18 (ANOVA) Reverse Pharmacognosy (RPG),  112, 113, Multivariate statistics,  213 117, 118, 125 N Root mean squared error or prediction National Center for Biotechnology (RMSEP), 225 Information (NCBI),  114 National Institutes of Health (NIH),  114 S Natural products,  38, 112, 134, 135, 143, 157, Scaffold,  34, 55, 135, 142–144, 166, 170 Score plot,  214, 220, 222, 224, 225 168, 170 Selnergy,  113, 118, 120, 125 as DPP-IV inhibitors,  192 Sensory data,  215, 217 databases, 196 of non-peptide nature,  192 CVA of,  220 PCA of,  218–220 SMILES fingerprints (SMIfp),  89–91

Index 251 SMILES representation,  89 Trace elemental sensory measurements of Statistical significance test,  215 wines, 215 Structural similarity,  3, 30, 32, 56, 115, 240 SuperScent,  83, 84 U SuperSweet,  83, 84, 86 Unsupervised methods,  31 Supervised methods,  23, 214 Svante Wold,  229 V Variance,  17, 19, 37, 38, 88, 214, 215, 222 T Tanimoto coefficient,  15, 16, 91, 167 Z Taste descriptors,  216 ZINC,  55, 84–87, 91, 93, 170 Topomer, 115